Corner entry tuning

[Ralph_]

Hey guys!

I’m just thinking about some „suspension stuff“ to tune a car for a better corner turn-in. I think I know how most of the cars setup affect different driving conditions (camber, toe, arb, rc, wtf, …) but I’m also confused at some points where I wanted to hear some different thoughts.

When I think about a car which points in very fast into a corner, I think about the following suspension tools which I can work with: toe, camber, RC (vertical, lateral), steering geometry (Ackermann), caster angle, springs, damper, ARB, inertia.. maybe I forgot something http://fsae.com/groupee_common/emoticons/icon_biggrin.gif

What I’m most confused of is the ARB. What I know is that, the stiffer the ARB is, the more weight transfer occurs on this axle, and the lower the lateral grip will be. This makes my car having much more understeer in the apex of the corner. But why does the car point in much faster(direct) into the corner?

My theory:

Building in an ARB means creating a direct connection between the corner inner and outer wheel. When I turn the wheels, some lateral acceleration will occur. This means, a weight transfer will result. What my springs, damper and ARB will see is the elastical weight transfer. When I use no ARB, the elastical weight transfer will go due the sprinsg and dampers. Depending on the roll inertia and speed and this stuff the car has some, I’ll call it “roll energy” which has to be converted into another form of energy until the chassis movement is completely “damped”. If you have no ARB, it will take some specific time until this energy has been converted. What you can see also in the OptimumG seminar folder(2009, page 233) is, that the EWTF takes some time and has a delay due to “bigger inertia”.

If I now install an ARB (or if a stiffen it), some of the EWTF can take “a faster way” from the inner to the outer wheel. This means less energy has to be taken by the springs and damper. So, the time of the EWTF will become less. And the dynamic corner weights will go up much faster ant don’t vary much.

But why does it make the car point in much faster? Due to the stiffer ARB, the outside wheels dynamic load goes up much faster. Which means, this tire heats up much faster and the overall lateral force at the front will go up much faster which in case yaws my car much faster.

When I increase the stiffness of the ARB in the front and rear the same amount, I do not increase weight transfer at the front, but the time the whole WTF takes will decrease?!

Maybe this question sounds stupid to some guys of you, but I don’t only want to know “Make the ARB stiffer! Why? Because we always did it like that!” and that kind of stuff. I would like to know what exactly happens in the car because of this change!

And/or is it about Inertia! Like. F – m*a = 0

The softer the whole suspension is, the longer the time until the system settles down is. Which means, if I have a soft suspension in roll and I steer my front tires into the corner, the chassis (inertia) will start to roll "to the corner outside". In a left corner, the chassis creates an inertia against the movement. If the suspension is too soft (the force is too small) the chassis will try to “bounce the front out of the corner”. If I stiffen the roll stiffness the reaction force is bigger, so the “bounce” is smaller which in fact makes my car turn in faster.

I try to validate my thoughts with an ADAMS model and some step steer simulation. But this shi.. beautiful program is really complicated and doesn't do what I want it to do! But as soon as I have some good reliable data, I will let you know.

Maybe some guys of you already worked on this problem and have some diagrams explaining it. Or maybe this problem has already been answered in a book which I don’t know or which I didn’t understand!

Cheers,
Ralph

UAS Graz

[Ralph_]

Hey guys!

I’m just thinking about some „suspension stuff“ to tune a car for a better corner turn-in. I think I know how most of the cars setup affect different driving conditions (camber, toe, arb, rc, wtf, …) but I’m also confused at some points where I wanted to hear some different thoughts.

When I think about a car which points in very fast into a corner, I think about the following suspension tools which I can work with: toe, camber, RC (vertical, lateral), steering geometry (Ackermann), caster angle, springs, damper, ARB, inertia.. maybe I forgot something http://fsae.com/groupee_common/emoticons/icon_biggrin.gif

What I’m most confused of is the ARB. What I know is that, the stiffer the ARB is, the more weight transfer occurs on this axle, and the lower the lateral grip will be. This makes my car having much more understeer in the apex of the corner. But why does the car point in much faster(direct) into the corner?

My theory:

Building in an ARB means creating a direct connection between the corner inner and outer wheel. When I turn the wheels, some lateral acceleration will occur. This means, a weight transfer will result. What my springs, damper and ARB will see is the elastical weight transfer. When I use no ARB, the elastical weight transfer will go due the sprinsg and dampers. Depending on the roll inertia and speed and this stuff the car has some, I’ll call it “roll energy” which has to be converted into another form of energy until the chassis movement is completely “damped”. If you have no ARB, it will take some specific time until this energy has been converted. What you can see also in the OptimumG seminar folder(2009, page 233) is, that the EWTF takes some time and has a delay due to “bigger inertia”.

If I now install an ARB (or if a stiffen it), some of the EWTF can take “a faster way” from the inner to the outer wheel. This means less energy has to be taken by the springs and damper. So, the time of the EWTF will become less. And the dynamic corner weights will go up much faster ant don’t vary much.

But why does it make the car point in much faster? Due to the stiffer ARB, the outside wheels dynamic load goes up much faster. Which means, this tire heats up much faster and the overall lateral force at the front will go up much faster which in case yaws my car much faster.

When I increase the stiffness of the ARB in the front and rear the same amount, I do not increase weight transfer at the front, but the time the whole WTF takes will decrease?!

Maybe this question sounds stupid to some guys of you, but I don’t only want to know “Make the ARB stiffer! Why? Because we always did it like that!” and that kind of stuff. I would like to know what exactly happens in the car because of this change!

And/or is it about Inertia! Like. F – m*a = 0

The softer the whole suspension is, the longer the time until the system settles down is. Which means, if I have a soft suspension in roll and I steer my front tires into the corner, the chassis (inertia) will start to roll "to the corner outside". In a left corner, the chassis creates an inertia against the movement. If the suspension is too soft (the force is too small) the chassis will try to “bounce the front out of the corner”. If I stiffen the roll stiffness the reaction force is bigger, so the “bounce” is smaller which in fact makes my car turn in faster.

I try to validate my thoughts with an ADAMS model and some step steer simulation. But this shi.. beautiful program is really complicated and doesn't do what I want it to do! But as soon as I have some good reliable data, I will let you know.

Maybe some guys of you already worked on this problem and have some diagrams explaining it. Or maybe this problem has already been answered in a book which I don’t know or which I didn’t understand!

Cheers,
Ralph

UAS Graz

[SNasello]

You are missing a very important part of the vehicle dynamics and that is the yaw moment and yaw inertia. For a fixed yaw a inertia, a higher yaw moment build up will result in a faster turn in. You need to look at what affects the yaw inertia at turn in. This includes things like weight distribution, wheelbase, tire cornering stiffness, and self aligning torque characteristics.

Creating a high yaw moment capacity at the front isn't so much the problem as having the yaw moment balanced at the limit. So you also need to look at yaw stability.

I would suggest looking at RCVD (CH.8) and working on a moment method simulation. It is much easier to run through different vehicle setups with such a simulation and see how it affects the shape of the CN-AY diagram. You can also make it as simple or complex as you like.

The other thing that must be clear is that the tire model is everything when it comes to vehicle dynamics simulations. Without a good, validated tire model, you will be chasing your tail. That means validating cornering stiffness as well as peak lateral force.

Remember springs, dampers and ARBs are only affecting one component of the tire characteristics, and that is the wheel load. This is definitely an important factor but is not the whole picture.

[Ralph_]

thanks for the fast reply!

I know that there is so much more behind the whole story.

But in this particular case I would like to know how the ARB effects the system. What's exactly happening when using the ARB to tune it.

When stiffening the ARB in the front and rear the same size, I do not change my WDB nor WB or the size of the wheel loads so in that case no cornering stiffness or SAT!?


I will look at RCVD again.. maybe i skipped something http://fsae.com/groupee_common/emoticons/icon_smile.gif

[Gruntguru]

What are you changing? Your first post suggests a stiffer front ARB. The last post suggests front and rear.

I don't understand how adding front ARB might improve turn-in. On the other hand adding front and rear will improve initial response in the same way that any increase in roll stiffness will do. One aspect of roll is movement of the CG towards the outside of the corner relative to the contact patches. This means the initial response to a steering input is a chsnge in direction for the contact patches that is not matched 100% by movement of the CG. So - the less roll, the quicker the initial response to steering input. I would suggest though, that this is a very brief delay and probably "feels" to the driver like a bigger effect than the actual effect in terms of lap times. In fact the extra stiffness can easily reduce grip to the extent that lap times suffer.

[Ralph_]

the post may be a little bit confusing.. but shouldn't it be regardless if I stiffen the ARB only the front or both ends of the car concerning turn-in.

If I increase my rear ARB, the value of the front anti roll moment stays the same. Just the distribution changes. What i want to say is, it doesn't matter if I change only front or both ARB. The effect of a better turn-in is the same.

The theory of the movement of the CG and the response time sounds good. But in case it's the same effect as I've written with my confusing energy-slow repsonse-reaction force-thing http://fsae.com/groupee_common/emoticons/icon_smile.gif

But what I learned, and also what you mentioned is the driver. Maybe the effect is very little, but the driver thinks he can go faster when the cars response is very fast. I think that's the reason why good drivers can go very fast even with setups which shouldn't work on the paper!

Like Carroll Smith once said:
"The sloppy sedan, even when being driven at its limit, gives the driver lots of time to make corrections to compensate for his errors in judgment. The Formula One car, driven as its limit, does not. This is, of course, why the star Grand Prix and Indy driver often don’t do very well at IROC – and why we used to see Jimmy Clark sometimes lose touring car races to drivers who couldn’t have come within ten seconds of his lap times in Formula One car. It is also the reason why there have been so many drivers who were brilliant in Touring and Grand Touring cars, but couldn’t get it done in real race cars."

[exFSAE]

1. ADAMS may not be doing what you want it to do, but I am pretty sure it is doing exactly what you *tell* it to do.

2. You are working under the presumption that adding a bunch of front bar DOES in fact "improve" turn-in. You may want to check & validate this assumption, particularly with data rather than driver comments.

3. You may learn a lot by making your own vehicle model outside of ADAMS, in Matlab for example. Run a model with and without roll inertia, or by sweeping inertia from effectively 0 to some high number.

Going along with point #3, if you want to make your car more responsive, all this stuff you're talking about with springs, bars, dampers, roll centers.. are all 2nd and 3rd order effects. Relatively small compared to the 1st order stuff, which are have not mentioned. Consider the case of a bicycle model - where is all the responsiveness and stability coming from? What parameter?

[RyMan]

To my understanding ARB's do not increase or decrease the amount of load transfer. They control chassis position relative to the wheels, so that as the outside wheel compresses the ARB lifts the inside wheel thus reducing the roll angle of the chassis. Along with the springs, they control the steady state behavior and roll angle of the vehicle. A stiffer bar/springs will mean less roll angle compared to a softer bar/springs for a given roll moment. Response as the driver interprets it is likely the amount and velocity of body roll as a function of steering input, which is what Gruntguru alluded to. So a chassis that rolls less and takes less time to reach the steady state roll angle seems more responsive. That can be achieved by stiffening up the springs and bars. The dampers apply force as a function the velocity of the wheel relative to the chassis (Or the chassis relative to the wheel)so they have an impact in transient situations where the positions of the objects relative to each other is changing with time, like corner entry.

As SNasello said above, yaw moment/yaw inertia=yaw acceleration. Yaw acceleration is essentially the turning response of the vehicle. You can teach a driver what real response is by having them drive karts, that way they aren't distracted by body roll. Hope that helped clarify things.

[Ralph_]

hey!
You are absolutely right with point 1http://fsae.com/groupee_common/emoticons/icon_biggrin.gif

2: Yeah, that's because I would like to know what's actually really going on. For example: If I make a technical analysis of something I can't write: Driver 1 says: I like it when the front is as hard as possible. and Driver 2 says: It feels better when it's not that hard.... what do i get from this quotes? Most drivers, especially in FSAE doesn't really know what they are talking about.

3. I already did it by changing the inertia in my ADAMS simulation.

I know that the most important part is the tire and so the cornering stiffness. But if I want to know what the ARB is really doing, I can't explain the influence of the CS?

cheers http://fsae.com/groupee_common/emoticons/icon_smile.gif

Add:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by RyMan:
So a chassis that rolls less and takes less time to reach the steady state roll angle seems more responsive.
</div></BLOCKQUOTE>

That's the point. Why does it make the car make more responsive? I think because a chassis which reaches the steady state roll angle faster, has much "smoother" corner weights. And a tire work at it's best when the load doesn't change that much.
(Isn't that what Renault told us with the mass and the vertical springs in the front and/or McLaren with it's Jdamper.. I know I'm talking about heave in this case.. but it's also about the wheel loads!)

[Rotary Sprocket]

[QUOTE]Originally posted by RyMan:
To my understanding ARB's do not increase or decrease the amount of load transfer. QUOTE]

That's how sway bars work. Stiffer bars create more weight transfer between the wheels and the opposite is true for a softer bar.

It is true that a stiffer platform will yield a more responsive vehicle but for a vehicle to have a good turn in response there needs to be more overall grip between the front tires than the rear tires.

There are many factors that affect the turn-in response of a vehicle; front and rear roll stiffness distribution, track widths, shock damping and even your steering geometry contributes to how well your vehicle turns in. The key is finding the right combination of all the variables.

[RyMan]

I'm just gonna go back and re-read/re-digest ch.18 of RCVD. Because I am confusing myself trying to visualize it.

[exFSAE]

Noodle on it. Draw it out. Sim it. Look at track data. Effects should be pretty clear.

This is a refreshing change-up from the "how du I design brk plz" threads.

[Goost]

This type of thread is what I come here looking for, but seem unfortunately uncommon.

ARBs change weight transfer Difference from front to rear, ie one pair transfers more though the total is nearly the same.

To add to SNasello's original reply I think he has a good point; yaw moment studies can teach you so much about what matters at corner entry and exit. For instance, I was always told that to make an oversteering car neutral, you either increase grip in the rear or decrease it in the front. my first thought is 'Why ever Decrease grip on one end? wouldn't that necessarily slow you down?' But RCVD and a little time with MATLAB I know the difference. the best thing is I can now study a corner as having five pieces, yawing entry apex exit deyawing, where the inertia And acceleration have an effect, the second order effects that were brought up. not only ARBs but some very interesting effects of changing the roll axis can be seen that way...

One thing I'm wondering is how does your simulation account for chassis stiffness? that may be where the solution is - er at least a way to see how much the difference in time between front and rear affects things. This may be incorrect, that's why I'm asking, but if the chassis is torqued as the front wants to roll and the rear doesn't, any change to the roll stiffness - especially in the rear - loses some effect, right? just a thought

[exFSAE]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">One thing I'm wondering is how does your simulation account for chassis stiffness? that may be where the solution is - er at least a way to see how much the difference in time between front and rear affects things. This may be incorrect, that's why I'm asking, but if the chassis is torqued as the front wants to roll and the rear doesn't, any change to the roll stiffness - especially in the rear - loses some effect, right? just a thought </div></BLOCKQUOTE>

I'd be wary of this thought process. Chassis rigidity isn't about the front wanting to roll and the rear not. They are not independent entities. You have lateral acceleration.. it acts on the mass distribution of the car, and you get a roll moment. How that roll moment is distributed front to rear is a function of the roll compliance you INTENTIONALLY design into the system (springs and bars), as well as the unintentional effects, e.g. frame rigidity.

If your frame / mechanical compliance is more significant than your springs and bars.. good luck tuning anything out.

[Ralph_]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Goost:
One thing I'm wondering is how does your simulation account for chassis stiffness? that may be where the solution is - er at least a way to see how much the difference in time between front and rear affects things. This may be incorrect, that's why I'm asking, but if the chassis is torqued as the front wants to roll and the rear doesn't, any change to the roll stiffness - especially in the rear - loses some effect, right? just a thought </div></BLOCKQUOTE>

My simulation uses a rigid chassis at the moment, because I don't have any idea how to create a flex body in ADAMS. This will take some time until I know it http://fsae.com/groupee_common/emoticons/icon_smile.gif. But there is a technical SAE paper (2000-01-3554) which describes the influence of the chassis roll stiffness into lateral load transfer.

But as you said, that's some point which really affects the turn in response.
If I think about it I would say, a chassis which is very flexible deforms a lot, thus the delay of the weight transfer takes much more time?! In this case, a soft front chassis reduces front lateral weight transfer. Which should increase front lateral force in the apex?! But due to the time delay and the feeling of the driver, it is not desirable for dynamic stability.

I really think it's because of the inertia of the car. When i turn left, the force created by the tires want to change my direction. But the cars inertia says "I want to go the same way as I do now!". The faster I can convert the energy, the faster the response will be and the better I will turn in.

I know my theory isn't common, as "rotary sprocket" already mentioned. But it sounds logical to me. http://fsae.com/groupee_common/emoticons/icon_biggrin.gif

But maybe its really all about the driver! http://fsae.com/groupee_common/emoticons/icon_smile.gif

[SNasello]

Ralph, modelling a flexible chassis in MBS does not require a complex FEM or ADAM/flex model. If you can't get your hands on one, model it a different way, such as a spring connecting the front and rear axle. Adding that compliance is better than not having it at all. When the frame is stiff enough then it might not have a huge effect, but it is also a good way to decide how stiff is stiff enough for your car.

There are only two things that you can do with springs and ARBs. One is to change the relative load transfer front to rear and the other is to increase the responsiveness. The ARB's don't cause the car to change direction though, the tires do. Springs, ARBs, dampers, camber, toe or whatever suspension adjustment you have are just tools to affect the tire forces and moments and the rate at which they build up. You need to examine this in every part of the corner, not just the apex. Tires are what produce your yaw moment and turn in response.

It is possible to create a car that turns in extremely quickly but cannot hold a very high peak lateral acceleration, because it has not mid corner stability. Then when you get on the gas on corner exit there is probably no chance at all of keeping the car on the track. The point is to find the balance between turn in response, mid corner lateral acceleration, and corner exit stability.

Once again, ADAMS may not be the best tool (or only tool required at least) for fully understanding this problem.

[Ralph_]

Good morning!

I think, or maybe I only hope, that ADAMS is one of the best tools at all to understand what changes in setup actually really do in the car. Most "static" analysis doesn't use a tire model like I do. Using different springs, ARB, RC and so on shows me whats happening with the corner weight and the tires lateral forces and moments...

Using a simple step steer simulation shows me very fast which setup really helps at corner entry and which one offers thee best mid corner lateral acceleration and yaw stability...
I see the results of a change in inertia or torsional stiffness of the car immediately. The hard part here is to explain why something is happening. Correct me if I'm wrong!

Today I will try to make sense of some setup changes. I will post some data which confirm my thoughts of changes, or maybe they will tell me I'm totally wrong http://fsae.com/groupee_common/emoticons/icon_biggrin.gif

[Rotary Sprocket]

You're right about ADAMS being an excellent tool for suspension tuning. I was pretty unfortunate to not have access to this program when I was designing our suspension so I spent most of my time reading up on suspension theory, banging through loads of static analysis and excel sheets and learning from trial and error.

Use the ADAMS simulations to get an idea for what set ups work the best and take the top 5 or so set-ups and try them in testing not only to see which one is faster but also which one your driver prefers.

Explaining why something is happening is hard because so many things are happening at once. Take advantage of the software and run lots of iterations on different variables. For example, raise and lower the front roll center height and see how it affects the car's handling. Then do the same for the rear roll center height. While doing this for different variables, you should start to see trends develop. Though it's tedious, getting an idea of how each variable affects the car's handling will make your job much easier when the car gets on the track for tuning.

[woodsy96]

I agree with Goost: threads like these are good and too far between.

I am going to go with agree with ex-FSAE's comment about driver feedback. But assuming the feedback is correct...

It could be the driver's preception of the car is "wallowy" (if that makes sense), becaue the car is rolling too much. Seems unlikely.

As everyone here has already said, the car's turn in response is generated from the control moment (turning the wheels). So I would start with what controls those forces/moments. F = function(Normal load, Slip Angle, Slip Ratio, camber angle). So trying to control these during corner entry will increase corner "turn in"

-the increased warp stiffness will increase the wedging effect-thereby reducing the yaw damping effect from the rear and increasing the control moment from the front simultaneously.

-Higher roll stiffness on the front also increases the total roll stiffness. This will increase the undamped natural frquency in roll (increased K) and simultaneouly reduce the damping ratio in roll (z = c/ c_crit), thereby increasing the damped natural frequency lots (w_d = w_n*sqrt(1-z^2). So faster weight transfer at both ends of the car.

-The car could be overdamped in roll across the front axle: therefore the damping balance is causing the front to have more LT during roll transients. Adding front roll stiffness increases totatl roll stiffness, transfering the load off the dampers onto the springs faster (basically very similar to the above point).

-Increased total roll stiffness reduces effects of roll steer due to lower roll angles (even during transients). This could be changing the effective ackermann at X roll angle.

-Increasing total roll stiffness reduces roll angles (even during transients), therefore the camber angles of the outside wheels (and maybe the inside) could be more favourable)

-Increased total roll stiffness places more energy from the overturning moment into displacing springs and less into moving dampers and accelerating roll inertias (M = Ia + cv + kx), thus the springs have a bigger effect on load transfer during the transient phase (and faster) than the inertia and dampers do. (Maybe, I probably need to think about this a bit more). If this was the case, then increasing rear roll stiffness would also increase turn in (but correspondingly more so).

/brain dump.

[Ralph_]

Hey guys!

it's been a while, but due to confusing results of my ADAMS model I had to work on the model again and again.. but now I think some of the results can be taken with good conscience http://fsae.com/groupee_common/emoticons/icon_smile.gif

The values may look different to fsae car values, that's may because it's a different(bigger) car.
No aerodynamic but drag included (no downforce)!

The graphs include 4 different lines (red = standard setup, blue = front arb, pink = rear arb, black = front and rear arb). The Simulation is a step steer simulation driving with 100km/h. After 1 second the steering wheel will be turned by 80 degrees in 0.1 seconds and stays the same value until the simulation ends.

First I looked at the vertical load of the front outer wheel.
http://i54.tinypic.com/nywbyb.jpg

What I see here is, the stiffer the front is, the faster the outer wheel gain load. If you stiffen the front and rear the same amount, the wheel load doesn't change mentionable.

http://i54.tinypic.com/w8rd45.jpg

The lateral load of the front outer wheel doesn't change in the transient part. The same slope and the same amplitude till 1.138 seconds. Later on, the car starts to stabilize itself and steady state starts. Here the grip drops by 60N when using only FARB. RARB increases front lateral force by 130N. Front and rear arb doesn't change steady state.

The overall front lateral force (inside and outside) also stays the same until 1.35 seconds. Afterwards the RARB shows higher grip at the front again.

http://i53.tinypic.com/qxmiyh.jpg

The yaw rate has the same slope and max. value at 41°/s at t=1.15. Afterwards it drops of with all setups. RARB shows higher yaw rate until the end of the simulation. That's because of the higher stabilizing moment of the front. Each setup shows a underdamped yaw movement.

http://i52.tinypic.com/9izdhg.jpg

What's about the "response" the driver feels? The roll movement is also underdamped. The Amplitude roll angle gets smaller with each setup. Using front and rear shows smallest roll angle. The Amplitude of the oscillation is smallest when using only RARB. The time until the system stabilizes itself stays the same. So the movement of the chassis stays the same.. I don't get what response really is.. http://fsae.com/groupee_common/emoticons/icon_biggrin.gif
But I think i should take the "definition" found in Milliken: it goes something like this: With response you talk about the time the car needs to reach steady state.

But in my simulation, the cars "response" is the same for each setup....


What I can see is, that the car doesn't point in much faster using the anti roll bar. The problem in this simulation is the most important part of the tire.. it's the temperature and also the pressure sensitivity. The Pacejka 89 coefficients/model doesn't use temperature or pressure changes. When you look at the normal load again, you see a much faster increase when stiffening the front. This, in case, heats up the front tire much faster. This means, the outer front wheel is able to develop it's max. lateral force much faster.

I never had the change to use tire surface temperature sensor on our car. Maybe someone of you used the sensors and can show (tell) us some values so that we can see how much difference in temperature we can expect. I know about a GT3 test where they implemented a temperature and pressure sensor inside the wheel (at the valve). 2 rounds later the sensor was dead because the temperature reached 130° inside and the sensor wasn't able to stand more. I know that GT3 is a much different race series, but the tires surface temperature is very high and changes so fast in each corner. If you are not able to reach the best temperature and pressure.. you will loose a lot of time.

I will do some more digging.. maybe I will be able to find more clear details which makes my confused mind clearer http://fsae.com/groupee_common/emoticons/icon_smile.gif

Cheers!

[Gruntguru]

The key thing demonstrated by your graphs is that changes to front and rear roll stiffness primarily affect steady state cornering rather than transients like turn-in. Clearly steady-state is the first handling characteristic that should be dialled in. To then make changes to transient behaviour without upsetting the steady state tune, you can look at parameters that are linked to inputs like steering angle (driver controllable) or higher order roll-stiffness parameters like damper settings.

Those associated with steering angle are particularly useful as they give the driver some control over the turn-in vs steady state characteristics for example he will probably like a setup that responds to an agressive initial input to the steering but is more "neutral" mid-turn with the steering a little straighter. Factors like caster (camber changes with steering input) or caster + scrub radius (jacking reduces front weight transfer with steering input) both have this effect. Some designers even offset the pushrod attachment to the upright rearwards to increase the latter effect.

[Silente]

This is a very interesting post.

One thing Ralph is not showing in his graphs is how camber is changing with different setups.

If roll angle increase/decrease very much with a certain setup, maybe also camber could be a lot different and tires could produce more/less cornering force.

Of course, adding temperature to the problem could make the analysis more complete.

[Chris B]

one thing that i'd like to chip in here with, not entirely related to the arb question but related to how it effects performance, ralph i think you mentioned earlier that FSAE drivers dont normally know what they're talking about. i dont know what kind of experience your drivers have but with my team we're lucky to have a few guys with experience in a few different types of racing so i normally take their feedback regarding set up above what should theoretically be the good set up (if that makes sense). my reason for this is that if the driver is comfortable, happy and confident in the car, they'll drive it as hard as the possibly can and lap times will reflect this. if they don't like the set up or don't feel confident that the car will dowhat they want it to then they won't be able to drive it as fast.

happy driver happy life.

by all means all the theory etc being discussed here is all very educational and ive really enjoyed reading it but don't discount the human element. for example we recently modified our steering that should have theoretically made the turning circle worse but the practical result on track was the opposite because it lightened the steering enough that our drivers could get to full lock much faster. this then gave them the confidence to drive hard and there was a pretty decent gain.

sorry for the essay. this is a really good thread though, but yeah with regards to set up, try and appease your drivers as it will benefit the performance of the car.

[ben]

Good post Chris. I've worked with some top level drivers and if you think they drive the same car the same way - think again.

We've had good success with OptimumG applying moment method techniques to car/tyre interaction and pretty much drive a lot of what we do now looking at Stability, Control and limit balance from a MMM N-Ay diagram. I'd say you're better off using that to setup your steady-state balance and general stability, then do transients on track with your dampers and driver.

As with all these things though, there's a theoretical (in the sim or textbook) result that is "best" but in reality if your driver doesn't get on with it then it will be slow. The product of setup potential and the driver's perception of it gives you your overall performance on track.

We've run different tyres on the same car for two drivers, one of whom hooked the front tyres the wrong side of the slip angle peak to make the car more comfortable to drive. You can argue that he's wrong, but if you can't replace him before the next session "theory" ain't a lot of help. Interestingly, would you go harder to deal with abuse if is the harder tyre gonna slide more and make the problem worse?

Stiff front bars and stiff dampers (i.e. digressive with a lot of low speed compression) make the car reactive to the driver - this might be illusory as exFSAE has mentioned, but if that makes him feel good, then you might gain lap time. On big GT cars and particularly LMP cars a stiff front bar can make the car pointy because the splitter rolls less - this could genuinely lead to higher front axle grip potential. The other factor already mentioned is camber - of you've got poor suspension geometry and/or crazy soft springs, then throwing bar at the front might control the tyre attitude better.

To sim all that properly in ADAMS or IPG would need a good damper model, a transient tyre model, which modeled Mx and Mz really well (which Pacejka doesn't by and large), a roll sensitive aero map and all manner of other things. I'd get in the right ballpark with a MMM sim and do transients on track.

Ben

[Tom Wettenhall]

I think I'm a bit below the general level of knowledge here, but I'll spit out what I know just in case, then I'll shut my trap.

I've been playing with this on track with the 2010 MUR car recently. My understanding is that, as you turn in, your unsprung mass begins to move, then the sprung mass responds with a magnitude dictated by the springs and a rate dictated by the dampers. Change that balance, like by adding front bar rate, and you change the response.

Fun experiment out on track: Try doing a few laps with full front low speed compression and no ARB, then full ARB and no LSC. Because there's no such thing as steady state, your instantaneous magnitude of weight transfer most of the way through a corner will be much the same, so your front-rear grip balance and therefore your max g and first-order handling characteristics will be much the same, but the transient response - turn-in, exit, the ubiquitous FSAE slaloms - will be markedly different.

Oh yeah, and your yaw inertia sets the ultimate limit on how quick your car will turn in. I'll be interested to see how far the current trend in huge front wings can be pushed before the old I_z starts to make itself felt.


I apologise if this is in fact all crap and I'm cluttering the thread with it.

[murpia]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by ben:
Good post Chris. I've worked with some top level drivers and if you think they drive the same car the same way - think again.

We've had good success with OptimumG applying moment method techniques to car/tyre interaction and pretty much drive a lot of what we do now looking at Stability, Control and limit balance from a MMM N-Ay diagram. I'd say you're better off using that to setup your steady-state balance and general stability, then do transients on track with your dampers and driver.

As with all these things though, there's a theoretical (in the sim or textbook) result that is "best" but in reality if your driver doesn't get on with it then it will be slow. The product of setup potential and the driver's perception of it gives you your overall performance on track.

We've run different tyres on the same car for two drivers, one of whom hooked the front tyres the wrong side of the slip angle peak to make the car more comfortable to drive. You can argue that he's wrong, but if you can't replace him before the next session "theory" ain't a lot of help. Interestingly, would you go harder to deal with abuse if is the harder tyre gonna slide more and make the problem worse?

Stiff front bars and stiff dampers (i.e. digressive with a lot of low speed compression) make the car reactive to the driver - this might be illusory as exFSAE has mentioned, but if that makes him feel good, then you might gain lap time. On big GT cars and particularly LMP cars a stiff front bar can make the car pointy because the splitter rolls less - this could genuinely lead to higher front axle grip potential. The other factor already mentioned is camber - of you've got poor suspension geometry and/or crazy soft springs, then throwing bar at the front might control the tyre attitude better.

To sim all that properly in ADAMS or IPG would need a good damper model, a transient tyre model, which modeled Mx and Mz really well (which Pacejka doesn't by and large), a roll sensitive aero map and all manner of other things. I'd get in the right ballpark with a MMM sim and do transients on track.

Ben </div></BLOCKQUOTE>
Thanks for the informative post Ben,

In 7 years of working with aero racecars I never came up with definitive answers to the questions posed here... My default assumption was always to try and work on the setup from an aero point of view - your comment on a stiff front bar moving aero balance forward under roll is very relevant here.

Tom's comment on yaw inertia is pertinent, I recall a test when nothing except yaw inertia was changed, by moving ballast. The higher yaw inertia setup 'understeered' which clearly indicated that driver did not get the car into what we might call a steady-state condition. And his style was more of the 'smooth' kind than the aggressive turn-in kind.

Also, I never resolved the front toe-in / toe-out vs. corner entry stability question. Toe-out was always used, and more toe-out would stabilise the car on corner entry under braking. But, we never worked out the theoretical basis for this, just observed it as a real effect.

Any ideas?

Regards, Ian

[Rotary Sprocket]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">
Also, I never resolved the front toe-in / toe-out vs. corner entry stability question. </div></BLOCKQUOTE>

I've always pondered this as well but have never come up with a reasonable answer. We've always run toe-out on the front because to our understanding it improved the turn-in of the vehicle. Toe-in was avoided because it improved straight line stability but we weren't all too concerned with that as navigating the corners was more important.

I've had the thought (and hopefully someone can confirm or deny this with some data) that it was better to run toe-in because on initial turn-in the outside tire doesn't encounter a point where the slip angle is zero causing a very small amount of time where your tire isn't developing lateral grip slowing the time it takes for the tire to reach max grip.

I don't know if that makes any sense, I've never been able to prove or dis-prove my theory. It's just an idea.

[murpia]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Rotary Sprocket:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">
Also, I never resolved the front toe-in / toe-out vs. corner entry stability question. </div></BLOCKQUOTE>

I've always pondered this as well but have never come up with a reasonable answer. We've always run toe-out on the front because to our understanding it improved the turn-in of the vehicle. Toe-in was avoided because it improved straight line stability but we weren't all too concerned with that as navigating the corners was more important.

I've had the thought (and hopefully someone can confirm or deny this with some data) that it was better to run toe-in because on initial turn-in the outside tire doesn't encounter a point where the slip angle is zero causing a very small amount of time where your tire isn't developing lateral grip slowing the time it takes for the tire to reach max grip.

I don't know if that makes any sense, I've never been able to prove or dis-prove my theory. It's just an idea. </div></BLOCKQUOTE>
Did you test whether toe-out improved turn-in, or just accept it 'should' be that way?

Your theory about running toe-in to improve turn-in sounds plausible also...

Undoubtably Ackermann has a part to play. Racecars with aero on 'large' tracks (unlike autocross) generally seem insensitive to Ackermann to the point where it gets ignored as a design parameter. Maybe it becomes important again on high-speed ovals to reduce tyre scrub? But, on an FSAE car a lot of Ackermann to get the inside front tyre generating a larger slip angle sooner could improve turn-in. That implies static toe-out also improves turn-in.

The best theoretical basis I could come up with for why front toe-out improves braking stability on 'big' cars has to do with bumps and load transfer. With front toe-out, if you hit a bump or the surface is not even when braking the increase in tyre contact patch load creates an increase in lateral force which opposes the increase in camber thrust from that tyre. With front toe-in the increase in lateral force adds to the camber thrust.

I would be very interested to hear any results from using ADAMS or other similar vehicle dynamics simulation to generate repeatable steering inputs (J-turns?) and to test the vehicle response with static toe-in and toe-out for various levels of anti & pro Ackermann.

Regards, Ian

[ben]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by murpia:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by ben:
Good post Chris. I've worked with some top level drivers and if you think they drive the same car the same way - think again.

We've had good success with OptimumG applying moment method techniques to car/tyre interaction and pretty much drive a lot of what we do now looking at Stability, Control and limit balance from a MMM N-Ay diagram. I'd say you're better off using that to setup your steady-state balance and general stability, then do transients on track with your dampers and driver.

As with all these things though, there's a theoretical (in the sim or textbook) result that is "best" but in reality if your driver doesn't get on with it then it will be slow. The product of setup potential and the driver's perception of it gives you your overall performance on track.

We've run different tyres on the same car for two drivers, one of whom hooked the front tyres the wrong side of the slip angle peak to make the car more comfortable to drive. You can argue that he's wrong, but if you can't replace him before the next session "theory" ain't a lot of help. Interestingly, would you go harder to deal with abuse if is the harder tyre gonna slide more and make the problem worse?

Stiff front bars and stiff dampers (i.e. digressive with a lot of low speed compression) make the car reactive to the driver - this might be illusory as exFSAE has mentioned, but if that makes him feel good, then you might gain lap time. On big GT cars and particularly LMP cars a stiff front bar can make the car pointy because the splitter rolls less - this could genuinely lead to higher front axle grip potential. The other factor already mentioned is camber - of you've got poor suspension geometry and/or crazy soft springs, then throwing bar at the front might control the tyre attitude better.

To sim all that properly in ADAMS or IPG would need a good damper model, a transient tyre model, which modeled Mx and Mz really well (which Pacejka doesn't by and large), a roll sensitive aero map and all manner of other things. I'd get in the right ballpark with a MMM sim and do transients on track.

Ben </div></BLOCKQUOTE>
Thanks for the informative post Ben,

In 7 years of working with aero racecars I never came up with definitive answers to the questions posed here... My default assumption was always to try and work on the setup from an aero point of view - your comment on a stiff front bar moving aero balance forward under roll is very relevant here.

Tom's comment on yaw inertia is pertinent, I recall a test when nothing except yaw inertia was changed, by moving ballast. The higher yaw inertia setup 'understeered' which clearly indicated that driver did not get the car into what we might call a steady-state condition. And his style was more of the 'smooth' kind than the aggressive turn-in kind.

Also, I never resolved the front toe-in / toe-out vs. corner entry stability question. Toe-out was always used, and more toe-out would stabilise the car on corner entry under braking. But, we never worked out the theoretical basis for this, just observed it as a real effect.

Any ideas?

Regards, Ian </div></BLOCKQUOTE>

Thanks Ian - glad some of me waffle made sense :-)

Interesting comment on the Yaw Inertia test. I've done stuff where a lower cornering stiffness tyre with the same ultimate peak Ay had "less grip" and I mean a lot as far as the driver was concerned.

So basically the driver cares about how quickly he can build angular acceleration early in the corner. If you give him less front cornering stiffness from the tyre or increase yaw inertia it's bad (unless you go to far - which I am assured is possible) and all of that's before the dampers have really had time to move.

So if you look at control moment from an MMM diagram and divide by yaw inertia that might be a good metric (yaw acceleration potential?) for transient behaviour without even doing a transient sim.

Ben

[Rotary Sprocket]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Did you test whether toe-out improved turn-in, or just accept it 'should' be that way?

Your theory about running toe-in to improve turn-in sounds plausible also...

Undoubtably Ackermann has a part to play. Racecars with aero on 'large' tracks (unlike autocross) generally seem insensitive to Ackermann to the point where it gets ignored as a design parameter. Maybe it becomes important again on high-speed ovals to reduce tyre scrub? But, on an FSAE car a lot of Ackermann to get the inside front tyre generating a larger slip angle sooner could improve turn-in. That implies static toe-out also improves turn-in.

The best theoretical basis I could come up with for why front toe-out improves braking stability on 'big' cars has to do with bumps and load transfer. With front toe-out, if you hit a bump or the surface is not even when braking the increase in tyre contact patch load creates an increase in lateral force which opposes the increase in camber thrust from that tyre. With front toe-in the increase in lateral force adds to the camber thrust.

I would be very interested to hear any results from using ADAMS or other similar vehicle dynamics simulation to generate repeatable steering inputs (J-turns?) and to test the vehicle response with static toe-in and toe-out for various levels of anti & pro Ackermann.

Regards, Ian </div></BLOCKQUOTE>

Unfortunately with all the things we tried to test this year that was one of the items that wasn't tested so it was just accepted that it "should" be that way.

With regards to Ackermann, I believe it plays a large part in the turn-in response of FSAE cars. I performed a parametric study using our tire data to determine what steering geometry would be appropriate for our car by evaluating the maximum cornering velocity at multiple radii and also what the vehicles angle relative to the radius of curvature (angle Beta) was for the given radius and cornering velocity. From these three variables (and also all the vehicles parameters; track width, car weight etc) the left and right steering angles could be determined. I picked two geometries that would satisfy most of the conditions and we tried them in testing. With the first geometry, the turn-in was ok but kind of sluggish. As our driver described to me, he had to turn out very briefly before he turned in to set himself up for a corner. After a day of testing we switched out the steering knuckles to try the other steering geometry and the change in handling was very apparent to him. Turn-in was very much improved so much so that when our driver took the first corner the way he had been used to with the other geometry his first input almost sent him off the track because the car responded so quickly. I think this was due to the fact that the the selected steering geometry allowed both front tires to reach their peak lateral grip at the same time where with the first geometry we ran one tire reached it's peak before the other causing a delayed response.

And Ackermann does play a part with aero cars on large tracks. I remember looking at a build sheet for a Lola Indy Car and they had four steering geometries to choose from ranging from pro-ackermann to anti-ackermann based on what type of track the car would be on.

[murpia]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Rotary Sprocket:
And Ackermann does play a part with aero cars on large tracks. I remember looking at a build sheet for a Lola Indy Car and they had four steering geometries to choose from ranging from pro-ackermann to anti-ackermann based on what type of track the car would be on. </div></BLOCKQUOTE>
I can certainly believe there are options available, (particularly with Lola). What I'm not convinced of is whether they have a significant enough effect to be worth tuning on road courses. Speedways I'm sure are different, but that's a scrub thing not a turn-in thing, like running very low rear toe-in at Monza.

Regards, Ian

[murpia]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by ben:
Thanks Ian - glad some of me waffle made sense :-)

Interesting comment on the Yaw Inertia test. I've done stuff where a lower cornering stiffness tyre with the same ultimate peak Ay had "less grip" and I mean a lot as far as the driver was concerned.

So basically the driver cares about how quickly he can build angular acceleration early in the corner. If you give him less front cornering stiffness from the tyre or increase yaw inertia it's bad (unless you go to far - which I am assured is possible) and all of that's before the dampers have really had time to move.

So if you look at control moment from an MMM diagram and divide by yaw inertia that might be a good metric (yaw acceleration potential?) for transient behaviour without even doing a transient sim.

Ben </div></BLOCKQUOTE>
I'm pretty convinced that's all correct. The 'best' racecars probably build yaw rate quickly but controllably, smoothly and stably. I.e. the drive can adjust the rate of increase of yaw rate with the steering wheel and as the weight transfer, aero and tyre temperature effects change the relative axle lateral grips those changes are predictable and smooth. Stability is the resistance of the car to disturbances such as bumps, road camber changes and surface grip variations. Good aero characteristics with steer and yaw is the winner here, even at the expense of a bit of peak downforce (but not too much).

A driver of such a car would probably report that he doesn't get any 'snaps' and that he can brake hard into the apex and get early on the power. He would probably also describe that as 'good grip' although I'd expect that if you could artificially reduce the lateral grip by e.g. 10% while retaining the good balance the loss of overall grip would be less obvious to the driver until he saw the laptime.

Steering weight (i.e. pneumatic trail feeback) is another thing drivers are sensitive to when defining 'grip'. If you weight up the steering with a geometry change, that can be reported as 'more grip' when in reality the tyres are performing very much the same. Kind of the inverse of the low cornering stiffness for the same lateral peak effect on driver feedback you described.


Changing the topic, what's your experience of the effect of inflation pressure on cornering stiffness? Is there a 'sweet spot' for maximum cornering stiffness with pressures either higher or lower than that having a cornering stiffness that falls away?

Regards, Ian

[ben]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by murpia:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by ben:
Thanks Ian - glad some of me waffle made sense :-)

Interesting comment on the Yaw Inertia test. I've done stuff where a lower cornering stiffness tyre with the same ultimate peak Ay had "less grip" and I mean a lot as far as the driver was concerned.

So basically the driver cares about how quickly he can build angular acceleration early in the corner. If you give him less front cornering stiffness from the tyre or increase yaw inertia it's bad (unless you go to far - which I am assured is possible) and all of that's before the dampers have really had time to move.

So if you look at control moment from an MMM diagram and divide by yaw inertia that might be a good metric (yaw acceleration potential?) for transient behaviour without even doing a transient sim.

Ben </div></BLOCKQUOTE>
I'm pretty convinced that's all correct. The 'best' racecars probably build yaw rate quickly but controllably, smoothly and stably. I.e. the drive can adjust the rate of increase of yaw rate with the steering wheel and as the weight transfer, aero and tyre temperature effects change the relative axle lateral grips those changes are predictable and smooth. Stability is the resistance of the car to disturbances such as bumps, road camber changes and surface grip variations. Good aero characteristics with steer and yaw is the winner here, even at the expense of a bit of peak downforce (but not too much).

A driver of such a car would probably report that he doesn't get any 'snaps' and that he can brake hard into the apex and get early on the power. He would probably also describe that as 'good grip' although I'd expect that if you could artificially reduce the lateral grip by e.g. 10% while retaining the good balance the loss of overall grip would be less obvious to the driver until he saw the laptime.

Steering weight (i.e. pneumatic trail feeback) is another thing drivers are sensitive to when defining 'grip'. If you weight up the steering with a geometry change, that can be reported as 'more grip' when in reality the tyres are performing very much the same. Kind of the inverse of the low cornering stiffness for the same lateral peak effect on driver feedback you described.


Changing the topic, what's your experience of the effect of inflation pressure on cornering stiffness? Is there a 'sweet spot' for maximum cornering stiffness with pressures either higher or lower than that having a cornering stiffness that falls away?

Regards, Ian </div></BLOCKQUOTE>

There's definitely a trade-off too high and you loose contact patch, too low you have reduced cornering stiffness and a lack of response. I've experienced both.

Ben

[murpia]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by ben:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by murpia:
Changing the topic, what's your experience of the effect of inflation pressure on cornering stiffness? Is there a 'sweet spot' for maximum cornering stiffness with pressures either higher or lower than that having a cornering stiffness that falls away? </div></BLOCKQUOTE>
There's definitely a trade-off too high and you loose contact patch, too low you have reduced cornering stiffness and a lack of response. I've experienced both.

Ben </div></BLOCKQUOTE>
What does the graph look like?

Is it a slower roll-off of cornering stiffness as inflation pressure increases and contact patch area decreases, compared to a sharper drop as the inflation pressure gets too low?

That's my 'hunch' but I have no data...

Regards, Ian

[dazz]

Such a great thread! Makes for a very engaging read.

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Rotary Sprocket:
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">
Also, I never resolved the front toe-in / toe-out vs. corner entry stability question. </div></BLOCKQUOTE>

I've always pondered this as well but have never come up with a reasonable answer. We've always run toe-out on the front because to our understanding it improved the turn-in of the vehicle. Toe-in was avoided because it improved straight line stability but we weren't all too concerned with that as navigating the corners was more important.

I've had the thought (and hopefully someone can confirm or deny this with some data) that it was better to run toe-in because on initial turn-in the outside tire doesn't encounter a point where the slip angle is zero causing a very small amount of time where your tire isn't developing lateral grip slowing the time it takes for the tire to reach max grip.

I don't know if that makes any sense, I've never been able to prove or dis-prove my theory. It's just an idea. </div></BLOCKQUOTE>

Something that sprang to mind with the above theory is that the apparent improved turn-in could be to do with how quickly yaw acceleration (not just lateral g) is generated with toe-out vs toe-in. And what effect the combined effects of yaw and lateral g have on your effective slip angles on the inside 'vs' outside. Also the rate at which yaw vs the rate at which lateral g is generated, ie yaw doesn't induce roll but lateral does, so if you toe setting generates yaw very quickly, and lateral g's more slowly, the car will feel like it darts into the turn and then takes a set as the lateral accel forces take over.

I've not had the time to get into vehicle sims so will have to ask here if I'm on the right track?

[Ralph_]

Just a few days away and there are a lot of really interesting posts.

A few words concerning my simulations. As I looked over and over it again, i have to say that this model includes static toe and camber.When eliminating all, the results change a little bit. I will doe some more simulation and post it here so that we can may discuss about it.

But as I already said, the model describes an FIA GT3 car. But I have no aero forces included. The experience I made at a GT3 race is, that a lot of changes which should ad more mechanical grip and may make the handling of an FSAE car without aero better, sometime results in worse grip when using a lot of aero, and vice versa.

My thoughts about a stiffer ARB making turn-in better stayed the same. When I look at the GFR-car, i think there's really nothing much more important than tire temp. So.. bringing temperature into it makes me faster. As far as i know, a stiffer ARB increases my natural frequency on the outer wheel, resulting in "less mechanical grip" but more heat in the tires.

What I also have to think about is the lateral RC migration (i read a lot about it here in this forum). If I have a lot of migration to the outside wheel in roll, the outside tire will go into less compression, making the wheel suspension "harder"(as optimumG calls it). So more tire temperature should come into the outside tires?!.. But that's just what the kinematic movement tells me. Maybe the force based lateral movement is much different... What also comes into play here is the MR. You can also make the outside with the MR ho harder or softer in roll? Which also brings temperature into account.

All in all, I think, there are so much variables in the suspension of the car which influence driving behavior.. Just fit the car to the driver and make him comfortable. As already someone said.. Make the driver happy and he will be fast http://fsae.com/groupee_common/emoticons/icon_smile.gif

@Ackermann and static-toe

Last year I used anti-ackermann geometry in combination with toe-out. I also used stiffer springs in front and rear because of a too low RC (under ground).. so too much roll occurred and I wanted to use as less ARB as possible.

This year we use pro-ackermann and tried both, static toe-in and out. The response for our (unexperienced) driver was always the same. It was better than last year, (using lower WR and no ARB).What we changed was the mechanical trail, so the steering effort is much smaller and now the drivers have more feeling in the steering wheel and they need less power to steer the car. So.. in that case I can not tell you what's really better. But.. i can try out some things in my simulation! Maybe it helps http://fsae.com/groupee_common/emoticons/icon_smile.gif


What I'm also in at the moment, concerning turn-in, is thinking about where the "instant" yaw axis is.
Depending on the inertia of the car, wheelbase, and weight distribution, your .. "center of percussion" varies. Depending on the things I listed.. the "yaw axis" for the instant moment where my front wheels start to steer, and the front lateral grip goes up, this center can be on the rear axis, in front of it or behind the axis. Depending on this location, the rear wheels will see a displacement and start to create a slip angle. This is one mayor part which really influences the turn-in behavior. But.. as i think.. it's hard to change.. just by putting parts(mass) somewhere else in the car. It justs tells you to put more weight in the middle of the car to get a better turn in. I just read about that in a german book, just have to look for the english one to post it.. (German book: Radführungen der Straßenfahrzeuge, Wolfgang Matschinsky, Seite 214)

@pressure and cornering stiffness

I have to say that I never focused on the difference of the CS by varying the pressure.
What I did was just looking that i have best pressure and so best tire "picture" on the car.

I think my brain is empty now. I've to reload it before telling some more wired things!

greets,
Ralph

[murpia]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Originally posted by Ralph_:
@pressure and cornering stiffness

I have to say that I never focused on the difference of the CS by varying the pressure.
What I did was just looking that i have best pressure and so best tire "picture" on the car. </div></BLOCKQUOTE>
Sometimes you haven't achieved the 'best' pressure when the car comes in and you need to make sense of the driver comments.

Knowing what this effect looks like should help this.

Regards, Ian